Question 6/6

"Fine wine while you dine"

Your solution will be scored on the correctness of your solution as well as on the methods you use.

You and your friend are hanging out at a fancy restaurant and begin talking about the quality of wines. You eventually decide that subjective discussions are of no consequence. You want to look at objective data to understand what is the best definition of wine quality.

Your friend is resourceful enough to find a dataset on wine quality assessments of various wine types. As you head back home your discussion becomes rather intense as to what attributes are important to the quality of wine.

So you pose the following questions with respect to wine quality:

1. Your friend argues that in red wine, pH has no impact on the quality of the wine. You say that is not true, who is right? Justify your answer.
2. You both agree that residual sugar of white wine will be related to the pH of the wine. How true is this? Justify your answer.
3. You decide there is no point arguing with each other and decide to build a model to predict the quality of red wine based on the various features that are available to you.
4. You have built a model but your friend is skeptical about the results. Convince your friend you’ve constructed a trustworthy model by discussing any metrics you used to test and validate the model.
5. You naively decide that your model should be able to predict the quality of white wine just as well. Check to see if this is true and justify your conclusion.

Task

In the Jupyter notebook carry out the above tasks and address accompanying questions

Data Dictionary

Number of Instances

• red wine - 1599
• white wine - 4898

Number of Columns: 12

Column information

• fixed acidity
• volatile acidity
• citric acid
• residual sugar
• chlorides
• free sulfur dioxide
• total sulfur dioxide
• density
• pH
• sulphates
• alcohol
• quality (score between 0 and 10)